[plug] date and time

[Craig Ringer]

> I got my post at the time I posted it.. I didn't realize so many of you
> stayed up so late ;)

Urrggh. I'm reading [plug] and other things while waiting for 120g of
data to copy off a failing HDD to a new one. Still at work at 00:02:43 -
yuck.

BTW, has anybody else noticed ridiclous rates of failures on Western
Digital HDDs, espcially 120g JBs bought fairly recently? I've had 2 fail
/this week/ , though the failures are progressive due to increasing
numbers of bad sectors. My home box's second drive is also dying, and I
had our gateway box die the harddisk death a few months ago, on an 80g
WD drive. The WD tools show the disks as being fine, but machines give
disk errors periodically and tend to crash, especially if swapping from
the disk that's failing. A S.M.A.R.T query reports bad sectors (numbers
vary from 2 to 440 across my failing drives), reallocated events, fatal
ATA errors, and often logged severe drive errors. The WD tools don't
seem to do anything more than a dumb disk scan, and since the sectors
have been reallocated (too bad about the data) and errors only generally
ocurr due to FS corruption from reallocated sectors or a new bad sector
being accessed, they don't tend to find anything wrong. 

The failure rates I'm seeing make me think "IBM DeathStar all over
again".

This is in several different machines, with several different chipsets,
PSUs, etc, on two sites. The older WD drives I have (1 year + ) seem to
be alright.

I'm replacing all the WD drives in my SATA RAID server with Seagate
Barracuda Vs, since I'm afraid that at the failure rate I'm seeing I'll
soon have 2 drives in that box fail at the same time. At least I keep
good backups - NBD + DDS4 = great whole-partition system images for
quick server rebuilds.

-- 
Craig Ringer
GPG Key Fingerprint: AF1C ABFE 7E64 E9C8 FC27  C16E D3CE CDC0 0E93 380D
	-- if it ain't broke, add features 'till it is. (or:)
	while (! broken) { features ++ ; broken = isBroken(features) }